Agile Robots Closes the thyssenkrupp Deal — Physical AI Just Swallowed 75 Years of German Engineering

01 — What Just Closed: The Deal in Plain Terms

The announcement came on April 1, 2026, from Munich. Agile Robots SE — a Physical AI company founded in 2018 as a spin-off from the German Aerospace Center — has successfully acquired the European and North American assets of thyssenkrupp Automation Engineering. The acquired unit will now operate as Krause Automation within the Agile Robots Group, preserving its operational identity while plugging into Agile's AI-native robotics platform.

The deal was first announced in November 2025 and cleared regulatory hurdles within months — unusually fast for a cross-border industrial transaction of this size. thyssenkrupp Automation Engineering's 2024 revenue was described as "in the hundreds of millions." Agile Robots itself reached approximately €200 million in revenue in 2024, having doubled revenue every year since founding. The combined group now spans more than 3,150 employees across over 10 new locations added through the acquisition, in addition to Agile's existing global footprint.

02 — Who Is Agile Robots? The Physical AI Stack Behind the Acquisition

Agile Robots is not a conventional robotics company. It was built by researchers — its founders came out of the German Aerospace Center (DLR), one of Europe's leading robotics research institutions — and its differentiator has always been its integrated approach to Physical AI: combining hardware, software, and real-world production data into a unified stack rather than selling isolated components or arms.

The company already has a broad portfolio. It makes intelligent robotic arms, autonomous mobile robots, and — launched just before the acquisition announcement — its first industrial humanoid, the Agile ONE. The Agile ONE is designed for material handling, pick-and-place, machine operation, and precise manipulation, with production at the company's own plant in Bavaria beginning in 2026. It will be showcased alongside Franka Robotics at Hannover Messe from April 20–24, 2026.

Agile Robots also maintains more than 1,000 AI and robotics experts on its R&D team — a figure that is extraordinary for a company of its age and scale. That research depth is precisely what attracted Google DeepMind as a partner and what makes the thyssenkrupp acquisition strategically coherent: Agile has the intelligence layer. What it needed was the implementation layer — the decades of project management, customer relationships, and system-integration expertise that industrial automation incumbents spend a generation building.

03 — What thyssenkrupp Brings: 75 Years, 650 Engineers, 10 New Locations

Founded in 1950, thyssenkrupp Automation Engineering — now Krause Automation — spent seven decades building one of Europe's most respected industrial automation integration businesses. Its specialty: turnkey automation systems for the most demanding manufacturing environments. Automotive body-in-white fabrication. Battery module assembly lines. Final assembly systems for major German OEMs. The kind of work where tolerances are measured in microns and a failed integration costs a customer millions per hour of downtime.

That customer base is what Agile gets with this deal. Not just 650 engineers — though that alone represents a 26% headcount expansion — but the relationships, the trust, and the multi-decade project history with the automotive and electronics manufacturers who will be the first to deploy humanoid robots at industrial scale. Krause Automation provides Agile with a direct pipeline into those decisions.

The strategic fit is tight. Agile wants to build full-stack Physical AI solutions: robot hardware, AI software, and complete plant integration, all from a single provider. Right now, most robotics companies do one or two of those well but not all three. By absorbing Krause Automation's system-integration expertise, Agile closes that gap. Customers who previously had to stitch together a robotics vendor, a software partner, and an automation integrator can now buy the complete package from Munich.

04 — Same Week, Two Moves: The Google DeepMind Partnership

The thyssenkrupp close did not happen in isolation. One week earlier, on March 24, Agile Robots announced a strategic research partnership with Google DeepMind to integrate Gemini Robotics foundation models into its hardware. The two announcements together — a hardware and integration acquisition plus a foundational AI model partnership — suggest a company executing on a very specific strategic roadmap rather than responding opportunistically to individual deals.

The DeepMind partnership plugs Gemini Robotics into Agile's physical platform — enabling robots trained on Google's large-scale AI models to operate on Agile's hardware in real industrial settings. Data collected during those deployments will, in turn, feed back into improving the underlying Gemini models. It's a data flywheel that benefits both parties: Google gets real-world industrial data at scale, and Agile gets access to foundation model capabilities that would cost hundreds of millions to develop independently.

Viewed together, the two moves trace the outlines of what Agile Robots is trying to become: the company that bridges the gap between cutting-edge AI research and the grounded, unglamorous, high-stakes reality of industrial automation — a gap that has historically been enormous and has prevented most AI-native robotics companies from achieving meaningful commercial scale.

05 — Why This Matters for the Global Humanoid Race

The conventional narrative of the humanoid robot race positions it as U.S. vs. China: Tesla and Figure on one side, Unitree and AgiBot on the other. Agile Robots complicates that picture usefully. It is a Chinese-founded company — Zhaopeng Chen was born in China and trained there before graduate work in Germany — that built its business in Munich, employs researchers from 60 countries, and has just acquired a heritage German industrial company. It is simultaneously backed by Google DeepMind and positioned as a champion of German engineering. The geopolitical categories are less clean in practice than they are in headlines.

More practically: the thyssenkrupp acquisition gives Agile Robots something that neither Unitree's open-source ecosystem nor Figure's BMW pilot quite achieves yet — a full-service, end-to-end automation partner that can walk a major manufacturer from early planning through robot hardware, AI software, and complete plant integration. That is the product that industrial buyers actually want. Not a robot. A solution.

For companies across the AI companion and high-fidelity humanoid interaction space — particularly those in Asia tracking where the physical AI infrastructure is heading — the Agile Robots trajectory offers a clear signal. The Physical AI era will not be won by hardware alone, or software alone, or even by the companies with the most advanced models. It will be won by companies that can close the loop between AI intelligence and industrial-grade execution. Agile, as of April 1, 2026, is the clearest current example of a company trying to do exactly that.